A Primer on Bayesian Decision Analysis With an Application to a Kidney Transplant Decision.
A Primer on Bayesian Decision Analysis With an Application to a Kidney Transplant Decision.
复制标题
DOI：
10.1097/tp.0000000000001145
复制
发表时间：
2016-03
期刊：
影响因子：
6.2
通讯作者：
Kaplan B
中科院分区：
文献类型：
--
作者：
Neapolitan R;Jiang X;Ladner DP;Kaplan B
为了提供个性化医疗，我们不仅必须确定最有可能对患者有效的治疗和其他决定，而且还要考虑患者在治疗的可能益处与可能的生活质量损失之间的权衡。有许多研究表明，各种治疗方法会对生活质量产生负面影响。即使我们拥有给定患者的所有信息，收集信息以做出决策，使决策对患者的效用最大化，也是一项艰巨的任务。临床决策支持系统（CDSS）是一种计算机程序，旨在帮助医疗保健专业人员进行决策任务。通过利用新兴的大型数据集，我们有望开发CDSS，可以预测治疗和其他决策如何影响结果。然而，我们需要超越这一点;即我们的CDSS需要考虑这些决定对生活质量的影响程度。这份手稿提供了一个介绍使用贝叶斯网络和影响图开发CDSS。这样的CDSS能够推荐最大化预测结果对患者的预期效用的决策。通过比较，我们研究了肾脏供体风险指数（KDRI）作为决策支持工具的好处和挑战，我们讨论了这个指数的几个困难。最重要的是，KDRI没有提供肾脏被接受时的预期生活质量与患者继续透析时的预期生活质量的衡量标准。最后，我们开发了一个模式的影响图模型的肾移植决策，并显示如何影响图的方法可以解决这些困难，并提供临床医生和潜在的移植受体一个有价值的决策支持工具。
To provide personalized medicine, we not only must determine the treatments and other decisions most likely to be effective for a patient, but also consider the patient’s tradeoff between possible benefits of therapy versus possible loss of quality of life. There are numerous studies indicating that various treatments can negatively affect quality of life. Even if we have all information available for a given patient, it is an arduous task to amass the information to reach a decision that maximizes the utility of the decision to the patient. A clinical decision support system (CDSS) is a computer program, which is designed to assist healthcare professionals with decision making tasks. By utilizing emerging large datasets, we hold promise for developing CDSSs that can predict how treatments and other decisions can affect outcomes. However, we need to go beyond that; namely our CDSS needs to account for the extent to which these decisions can affect quality of life. This manuscript provides an introduction to developing CDSSs using Bayesian networks and influence diagrams. Such CDSSs are able to recommend decisions that maximize the expected utility of the predicted outcomes to the patient. By way of comparison, we examine the benefit and challenges of the Kidney Donor Risk Index (KDRI) as a decision support tool, and we discuss several difficulties with this index. Most importantly, the KDRI does not provide a measure of the expected quality of life if the kidney is accepted versus the expected quality of life if the patient stays on dialysis. Finally, we develop a schema for an influence diagram that models the kidney transplant decision, and show how the influence diagram approach can resolve these difficulties and provide the clinician and the potential transplant recipient with a valuable decision support tool.